Viking Archaeology, Sagas, and Interdisciplinary Research in Iceland’S Mosfell Valley

research in action

Report from the Field

Viking Archaeology, Sagas, and Interdisciplinary Research in Iceland’s Mosfell Valley

Jesse Byock 1 and Davide Zori 2

This article is dedicated to the memory of Phillip Walker, By applying different sources and research our friend, University of California colleague, and co- techniques—archaeological, scientific, and human- director of the Mosfell Archaeological Project istic—MAP is constructing a picture of habitation and environmental change in the Mosfell Valley over Working in the glaciated and once-wooded Mosfell the course of the Old Icelandic Free State, a Viking Valley (Mosfellsdalur) in southwestern Iceland, the Age parliamentary state that continued until the Mosfell Archaeological Project (MAP) is unearth- mid-thirteenth century (Fig. 1; Byock 2001). MAP’s ing the prehistory and early settlement history of that region during the Viking Age (ninth to eleventh century C.E.).3 This article offers an overview of 1 Scandinavian Section, UCLA and Cotsen Institute of Archaeology, MAP’s recent archaeological research at the farm of UCLA. Hrísbrú in the valley. This site was the home of the 2 UCLA Center for Medieval and Renaissance Studies; Icelandic Centre Mosfell chieftains (the Mosfellsdælingar, “the People for Research (Rannís). of Mosfell’s Dale”), a powerful Viking Age family 3 The Mosfell Archaeology Project (MAP) is directed by Jesse Byock, UCLA professor of Old Norse and Archaeology. MAP’s field director is of leaders, warriors, farmers, and legal specialists. Davide Zori, UCLA Cotsen Institute of Archaeology Ph.D., 2010. A joint Within the context of MAP’s findings, this article Icelandic and American project, MAP is interdisciplinary and international considers the relationship between modern archaeo- with archaeologists, scholars, and students from different countries and universities. For further information, see the forthcoming book, Viking logical methods and Iceland’s medieval writings, Archaeology in Iceland: The Mosfell Archaeological Project, ed. Davide Zori especially the family sagas. and Jesse Byock (Turnhout: Brepols Publishers, 2014).

124 | backdirt 2013 Figure 1. Map of the Mosfell Valley region. The red squares show the locations of MAP’s excavation and survey sites. The entrance to the valley lies between Mosfell Mountain to the north and Helgafell Mountain to the south. The farmstead of Hrísbrú sits on the lower slopes of Mosfell Mountain from where it controlled the entrance to the valley and looked down on the Viking Age port at Skiphóll (Ship Knoll) in Leiruvogur Bay.

excavations at Hrísbrú have revealed a large longhouse, a timber/stave church from the conversion- age transition from paganism (ca. 1000 C.E.), an early graveyard with mixed pagan and Christian attributes, and a pagan cremation burial site (Figs. 2, 3, and 4). Together these exceptionally well-preserved remains form the core features of a chieftain’s high-status farmstead. At this time, we are research- ing and excavating fourteen sites throughout the valley (see Fig. 1). These excavations are providing a detailed picture of Viking Age life in Iceland and in the North Atlantic. The sites include monumen- tal stone ship settings—stones arrayed to form the outline of a ship—at the inland end of the valley (Fig. 5), and a Viking Age port at the valley’s coastal Figure 2. The modern farm of Hrísbrú showing the locations of mouth at Leiruvogur Bay, where the rivers of the the major excavation areas of Kirkjuhóll (Church Knoll), Hulduhóll valley flow into the sea.4 The extensive assemblage of (Elven Hill), and the tún (the ancient homefield) mentioned in this sites in the Mosfell Valley forms a powerful political, article. The old road leading past Hrísbrú lay between the Christian religious, and governmental landscape. church on one side and the pagan burial mound on the other. The Mosfell Valley is the type of Icelandic community of the Viking Age that produced the Icelandic sagas, one of the world’s great bodies of

4 Leiruvogur, “Tidal Flats Bay,” is sometimes spelled “Leirvogur” and may also be translated as “Clay Bay.” Both translations make sense as the bay is still today both clayey and muddy.

backdirt 2013 | 125 Figure 3. Site plan of the excavated Viking Age chieftain’s farm at Hrísbrú. The map shows the Viking Age longhouse and the conversion- era church (ca. C.E. 1000) with a surrounding burial yard.

literature. Most archaeologists working in Iceland roads, burials, agricultural enclosures, and port today avoid the sagas, dismissing them as fictitious facilities before they are destroyed by modern writings. We take a different view. We employ construction. Iceland’s medieval writings as one of many datasets MAP is highly interdisciplinary in its archaeolog- in our excavations, and the archaeological remains ical approach, using the tools of archaeology, history, that we are excavating in the Mosfell Valley appear anthropology, forensics, environmental sciences, to verify our method. Together, the written medieval and saga studies in a coordinated methodology. We sources and the archaeological discoveries offer new call our research framework Valley System Archae- information about Iceland’s earliest past and about ology, a concept we developed and which we find the Viking Age in general. well-suited to Viking and North Atlantic archaeology. MAP is also an example of archaeology in transi- Guided by this framework, we combine analyses of tion. When we started in 1995, the valley was rural. the Mosfell Valley’s cultural, scientific, and envi- It lay beyond the outskirts of Reykjavík, Iceland’s ronmental landscapes from the coastal regions at capital. But the situation has changed: the Mosfell the western lowland mouth of the valley up into the Valley is rapidly becoming part of sprawling greater highland heaths that rise at the valley’s eastern end. Reykjavík, and the Viking Age sites are now threat- The Mosfell region is a suitable test case for the ened by urban development. Since we started our utility of the Valley System concept. The valley, the excavations, a large area of the Viking port area at surrounding highlands, and the lowland coastal Leiruvogur Bay has been turned into paved housing areas are an interlocking system of natural and man- subsections. Where earlier we herded horses and made components. The area encapsulates the major sheep off our sites, we now compete with bulldoz- ecologies of Iceland: coastal, riverine, and highlands ers and larger machinery. With the swift advance of with volcanic soil; it contains rich archaeological urbanization, our task is to find and document the heritage, connected to a broad collection of oral archaeological remains of turf buildings, ancient narrative and medieval writings. The time-frame of MAP’s research begins with Iceland’s earliest

126 | backdirt 2013 research in action

Figure 4. Architectural drawing of the buildings at Hrísbrú in the Mosfell Valley reconstructed from the MAP site plans. The church, which is of timber/stave construction, is approximately 8 meters from the turf-clad longhouse.

Figure 5. A ship setting found in the Mosfell Valley. Such man-made stone settings arrayed in the shapes of ships are widely distributed throughout mainland Scandinavia and are often connected with the rituals of death. This is the first such monument to be found in Iceland. The ship setting, 30 meters long and 10 meters wide, is made up of 69 laid stones.

backdirt 2013 | 127 locate turf structures (Bathurst et al. 2010). We are analyzing architectural techniques (Byock and Zori in press); trade and exchange as witnessed by the artifact record (Hansen et al. in press); the production, forms, and uses of iron (Zori 2007); usage of smaller activity areas, such as the sel or summer dairy stations; roads and paths (Connors in press); the intra-site artifact distribution patterns in a Viking longhouse (Milek et al. in press); and the role of feasting in an environmentally marginal North Atlantic society (Zori et al. 2013). MAP’s archaeological findings, including our artifact collection, indicate that the region is cultur- ally representative of early Iceland. The community that evolved in the Mosfell Valley was in many ways a self-contained social and economic unit. It was also connected to the rest of Iceland through a Figure 6. Four “eye beads,” originally from the eastern shores of network of roads. Two major east–west roads lead the Caspian Sea, were found hidden in a pit dug in the floor of the Hrísbrú longhouse. More than 30 beads were found within the through the valley, one on the north slopes of the longhouse, the largest number of such finds recovered within an valley and the other on the south side. These routes Icelandic longhouse of the Viking Age. Beads were valuable trade connect into major north-south routes. They also goods in the Viking Age, and these finds are consistent with the lead to nearby Thingvellir (the “thing or assembly wealth and status of the inhabitants of the Hrísbrú farmstead. plains”), the meeting place of the annual Viking Age parliament, the Althing, thirty kilometers (about eighteen miles) to the east of the Mosfell Valley. The artifacts, including imported glass beads from ninth-century settlement period, called the landnám the Mediterranean and Central Asia, show that the or “landtaking,” when Norse and Celtic settlers first Mosfell Valley was also in contact with the inter- sailed from northern Europe to uninhabited Iceland. national trade and travel of the Viking Age (Fig. The time frame of our current excavations continues 6). Leiruvogur Port at the western mouth of the through the transformation from paganism to Chris- valley provided commercial and cultural contact tianity and into the thirteenth century C.E. with the larger Scandinavian and European worlds. The specialists on the MAP team explore, among The connections went possibly as far as Constan- other subjects, local power structures in wider tinople (connected historically to Scandinavia and Iceland contexts (Byock in press; Byock and Zori in Iceland through the Varangian Guard, the imperial press); the changes over time in subsistence strate- bodyguard of the Byzantine Emperor, composed of gies (Erlendsson et al. in press; Zori et al. in press); Northmen), the Caspian Sea, and Greenland. health and disease (Holck in press; Eng in press); MAP’s archaeological work began in the mid- the place of origin of the Norse immigrants (Grimes 1990s with background research on historical et al. in press); the role of Leiruvogur Harbor in the sources, field surveys, and test excavations (pub- international exchange of the Viking Age (Hilberg lished in Byock 1993, 1994a, 1995; Earle et al. 1995). and Kalmring in press); the integration of texts Major excavations began in 2001, and subsequent and archaeology (Erlandson et al. in press; Byock field seasons have documented a rich Viking Age in press; Byock et al 2005); the organization of the and landnám occupational history (Byock et al. 2005; local settlement pattern (Zori in press); cooperation Holck 2005; Byock 2009). The 2001–2004 excava- between local municipalities and archaeologists tions at Hrísbrú in the valley revealed the presence (Thórðarson in press); the paleobotany of early Ice- of significant remains, including an early church, landic settlements (Martin in press) and the develop- a surrounding cemetery, and an adjacent burial ment of innovative subsurface survey methods to mound containing human cremation remains (Fig. 7). Our subsequent field seasons have expanded the

128 | backdirt 2013 research in action

Figure 7. The conversion-age timber/stave church at Hrísbrú. The plan shows the two parts of the building, the nave on the left and the chancel on the right. By the end of MAP’s graveyard campaign, we had found the remains of 26 burials. A number of the burials appear to have been emptied. Burial Features 4 and 46 on the north and south side of the chancel are reburials containing disarticulated skeletons (see Fig. 14). These bones were moved to the churchyard from earlier graves, probably pagan. scope of the work, and in recent years, we excavated 10). The Icelandic Book of Settlements or Landnámabók at Hrísbrú a large (28 m long) early tenth-century preserves the tradition that Skeggjastaðir was the longhouse (Figs. 8, 9, and 16). In Old Norse/Ice- first farm in the valley, established by and named for landic, this type of longhouse is called an eldskáli or Thord Skeggi, the first Viking Age settler to claim firehall, named for the long fire (langeldr) down the land in the region. Our work at Skeggjastaðir is center of the hall. In the case of the Hrísbrú long- just beginning and promises to be another example house, the long fire is 5.8 meters long, making it one of fruitful interdisciplinary research (see Zori in of the largest such hearths excavated in Iceland. press). Two additional sites discussed in this article Our current excavations on the Hrísbrú farm are ship-like stone settings in the low highlands on focus on three areas: Kirkjuhóll (Church Knoll), a the eastern inland end of the valley and the port at hillock just behind the modern farm’s stable; the tún Leiruvogur at the western end of the valley. or hayfield just north of Kirkjuhóll; and Hulduhóll The stone ship settings in the eastern end of (Elfin Hill, Knoll of the Hidden People), the hillock the valley are unusual in Iceland. To date, no other located about 20 meters west of Kirkjuhóll (see Fig. examples of such monuments have been found on 2). Numerous additional sites are under investiga- the island. The ship settings appear to be remnants tion in the valley. At Skeggjastaðir, in the eastern of the pre-Christian ritual landscape. Probably they end of the valley, our recent campaign of subsurface are mortuary sites, but they could also be assembly archaeological testing has located the remains of a sites, and the one possibility does not preclude the previously unknown Viking-period farmstead (Fig.

backdirt 2013 | 129 Figure 8. The longhouse at Hrísbrú in the Mosfell Valley was constructed around the year 900. This picture was taken during MAP’s excavation, when the long fire down the center of the sunken central hall was just coming into view. Also coming into view are the raised benches along the sides of the central hall on which people ate and slept. At this point in the excavation, the benches were still partially covered by toppled small stones from the building’s side walls. These stones from the collapsed walls were removed during subsequent excavations, revealing the full extent of the meter-wide benches. The door through wall on the left was large enough for livestock. The smaller entrance on the other end of the same wall was the main human entrance. This door looked directly down on the Viking Age harbor at Leiruvogur.

other. Indeed the Mosfell Valley stone ships are stone features are not recorded in any medieval or comparable to ship settings found throughout the modern written sources, and they came as a surprise Viking world, which are connected with both burial to individuals and families who had lived in the and assembly sites. The largest Mosfell setting is valley for generations. The placement of the ship set- approximately 30 meters long and 9 meters wide, tings at what we believe is the possible boundary of typical proportions of a Viking Age ship (see Fig. 5). the old Mosfell farm is consistent with the location The long axis is oriented east–west, with the prow of burial mounds on land boundaries elsewhere in pointing west toward Leiruvogur Bay and the sea. Iceland (Zori 2010; on pagan burials in Iceland, see MAP’s excavations, especially of the stones at the Fríðriksson 2009). prow and in the center, have determined that the From the start of our research, we have incor- ship-settings predate the erosion in the area. These porated ethnographic components. In particular,

130 | backdirt 2013 research in action

Figure 9. Structural view of the Hrísbrú longhouse detailing the building’s internal wooden frame. The external turf walls are indicated by the dotted lines around the building. The drawing shows a second- story loft to the east of the main entrance and partially over the central hall. Note the two entrances on either end of the building.

Figure 10. Map of Skeggjastaðir showing the location of the medieval farmstead discovered with subsurface coring. According to the tradition retained in the medieval Icelandic Book of Settlements (Landnámabók), this farm was settled by and named after the valley’s first settler, Thord Skeggi.

backdirt 2013 | 131 we pay careful attention to the oral histories of local and after the conversion to Christianity (ca. 1000), families. The farmers at Hrísbrú—Ólafur Ingimun- Grím Svertingsson, Law Speaker of Iceland from darson, his son Andrés Ólafsson and their fami- 1002 to 1004, lived at Hrísbrú (Old Mosfell). Egil’s lies—are extremely knowledgeable about their valley Saga, Gunnlaug’s Saga (Gunnlaugs saga ormstungu), (see Fig. 17). Their family has lived on the land for and Hallfred’s Saga (Hallfreðar saga) offer detailed many generations, and Ólafur and Andrés spent a information about the Mosfell chieftaincy (goðorð). great deal of time with us explaining the landscape. According to these sagas, the Mosfell chieftains When we began excavating in the Mosfell Valley controlled the area known as Nesin (the Nesses), the in 1995, the two adjacent knolls of Kirkjuhóll and headlands or promontories of the southern coastal Hulduhóll were used as pasture. They were covered region that stretched out from the valley’s mouth with grass, and their surfaces were undisturbed past modern Reykjavík and farther west. In a direct except where the trampling of cows exposed small line of sight from the front door of the Hrísbrú patches of earth. No agricultural machinery was ever longhouse, one sees all of Reykjavík and beyond to used on Kirkjuhóll because of the reverence attached Seltjarnarnes (Fig. 11). Conversely, standing in the to the knoll in oral memory as the site of an ancient old harbor in the center of old Reykjavík and looking church. This situation is fortunate since most con- east, one sees the farm of Hrísbrú on the slopes of temporary Icelandic farms are highly mechanized. Mosfell Mountain. Hulduhóll, the site of the cremation burial, had In the medieval sources, the Mosfell chieftains also been spared the effects of agricultural machin- are said to have called up armed men from the ery. Stories attached an interdiction or taboo to Nesses to support them in times of armed conflict. Hulduhóll to leave it alone, because it was inhabited So too the Mosfell chiefs are said to have entered by “the hidden people” or elves, who were dangerous into marriage alliances with the chiefs at Borg (Fig. if disturbed. As it turned out, oral memory proved 12). This powerful family was descended from the strong. Both knolls were connected with ancient Norwegian Skalla-Grím Kveldúlfsson, the first set- mortuary rites. They contained human remains tler or landnámsmaðr in Borgarfjörður and father from the Viking Age, both Christian and pagan. of the Viking Egil from Egil’s Saga. Given that Borg In addition to modern oral memory, the archae- controlled the coast of a fjord area, a day’s ride to ology of the Mosfell Valley is aided by a wealth of the north of Mosfell, alliance between these two surviving medieval Icelandic writings describing chieftain families would have been quite logical and the valley’s sites and people. Listed in the notes, this would have bolstered the power and authority of extraordinarily large and varied collection of writ- both families. The two were close enough to support ings are rich sources about the Mosfell chieftains.5 each other, but far enough away not to compete for They make the Mosfell Valley an ideal test case for thingmen (followers) or natural resources. reconsidering the validity of the sagas as historical As archaeologists piecing together the different and archaeological sources. The writings tell that possibilities of the geographical and political land- the geographical position of the chieftains’ lands scapes, we might have speculated on the likelihood and their area of power allowed them to monitor of an alliance between these families. Iceland’s sagas and benefit from the travel and trade that passed made such speculation unnecessary. The following through their port and valley. Egil’s Saga (Egils saga textual example (chapter 81 of Egil’s Saga, written Skalla-Grímssonar) and The Book of the Icelanders around the year 1220) offers insight into the func- (Íslendingabók) recount that in the years before tioning of an ancient alliance between Borg and Mosfell. Egil, for example, was originally from Borg. In his later years, he gave his chieftaincy at Borg to 5 These include The Book of Settlements (Landnámabók); The Book of the Icelanders (Íslendingabók); Egil’s Saga (Egils saga Skallagrímssonar); The Saga his son Thorstein and moved to Mosfell. of Gunnlaug Serpent-Tongue (Gunnlaugs saga ormstungu); Hallfred’s Saga An episode from Egil’s Saga makes clear that the (Hallfreðar saga); The Saga of the People of Kjalarness (Kjalnesinga saga); The connection between the chieftains at Mosfell and Saga of the People of Floi Bay (Flóamanna saga); The Tale of Thorstein Bull-Leg (Þórsteins þáttr uxafóts); Njál’s Saga (Njáls saga); The List of Priests (Pres- Borg remained strong. According to this text, after tatal); The Saga of Thorgils and Hafliði (Þorgils saga ok Hafliða); and The Egil departed for Mosfell, his son, Thorstein at Borg, Short Saga of Orm Storolfsson (Orms þáttr Stórólfssonar) in Flateyjarbók. found himself in a property dispute with his neighbor. This conflict was especially dangerous, because

132 | backdirt 2013 research in action

the neighbor had secured the alliance of several chieftains, all of whom stood to profit if Thorstein were to lose the dispute. The matter resulted in a showdown of force at the local springtime assembly, the várþing, where Thorstein found himself outnum- bered. If matters went against him, he stood to lose his lands, his chieftaincy, and perhaps his life. The saga describes the dramatic resolution of this crisis at the Borgarfjörður assembly (called the “thing”) as follows: That day men went to the thing slope and discussed their lawsuits, for in the evening the courts would convene to consider prosecu- tions. Thorstein was there with his following and had the greatest say in the conducting of Figure 11.The Nesses (headlands) with the Mosfell Valley in the the thing, because that had been the custom center. The people of the Nesses are referred to in the sagas as the Nesjamenn (The Men of the Nesses). The farm at Hrísbrú over- while Egil was still a leader and was in charge looks the Nesses, which can also be seen from the main entrance of the chieftaincy. Both sides were fully of the longhouse. armed. From the thing site, men saw a group of horsemen come riding up along the Gljúf River. Their shields shone in the sun and there in the lead, as they came toward the spring assembly, was a man in a blue cape. On his head was a gilded helmet and at his side was a shield worked with gold. He held in his hand a barbed spear, its socket inlaid with gold. A sword was bound to his waist. Egil Skalla-Grímsson had come with eighty men, all well-armed, as if ready for battle. It was a carefully chosen troop. Egil had with him the best farmers’ sons from the Nesses to the south, those whom he thought the toughest fighters.6 After his timely journey to Borg, Egil returned to Mosfell. Years later when he died (ca. 990) in the Mosfell Valley, Egil was first interred in a pagan burial mound. Later, after Iceland had converted Figure 12. Southwest Iceland, showing the two allied centers of to Christianity, he was reburied twice in Christian chiefly power, Borg in Borgarfjörður and Hrísbrú (Old Mosfell) in graveyards. His second burial (first reburial) was at the Mosfell Valley. Hrísbrú (Old Mosfell) in the early eleventh century. His third burial (second reburial) was at Mosfell (New Mosfell) in the mid-twelfth century (see Byock 1995). But for the sagas, we would know nothing of Egil’s posthumous travels, which provide a wealth of information about the local solutions to religious changes from paganism to Christianity. 6 Translation by Byock. The line about the Nesses reads in Old Icelandic: “hafði Egill haft með sér ina beztu bóndasonu af Nesjum.”

backdirt 2013 | 133 The sagas also give us a good deal of information In the summer Hallfred sailed out [from about the other sites in the region. For instance, the Norway] to Iceland, landing his ship in Leiru- Leiruvogur Port, at the intersection of the Nesses vogur, south below the Mosfell heath. At the and the Mosfell Valley, is mentioned in more Ice- time Önund was living at Mosfell. Hallfred landic sagas than any other early harbor in this part was required to pay half a mark of silver to of the island. According to The Short Saga of Orm Önund’s house servant, but refused harshly. Storolfsson, (Orms þáttur Stórolfssonar), Hallfred’s The servant came home and told of his Saga and The Saga of Gunnlaug Serpent-Tongue, ships trouble. Hrafn [Önund’s son] said it was to be from Norway landed in Leiruvogur. In The Saga of expected that the servant would get the lower the People of Floi Bay (Flóamanna saga), a man called part of the bargain in an exchange between Thorgils intends to leave Iceland and establish a new them. And in the morning, Hrafn himself farm in Greenland. He goes to Leiruvogur Harbor rode to the ship, intending to cut the anchor to purchase an ocean-going ship although other har- cable [causing the ship to drift and get stuck bors are closer to his home (Zori 2010, 183). MAP’s on the mud flats of Leiruvogur (Clay Bay)] to archaeological coring, geophysical testing, and sur- make sure that Hallfred and his men did not vey of Leiruvogur Port has shown why Thorgils went leave. Then men intervened between them there to find a ship (Fig. 13). and took part in reconciling them. The result Leiruvogur Bay, with its highly sheltered anchor- was that Hallfred paid half again more than age, offers more protection for anchored or landed the servant had demanded. With this they ships than any other harbor in this part of Iceland. parted. The bay reaches far inland, and its bays, estuary, (Byock trans.) and inner lagoon are protected behind an unusual combination of natural barriers. These include an Ok at sumri fór Hallfreðr út til Íslands ok extensive series of small islands and nesses (prom- kom skipi sínu í Leiruvág fyrir sunnan land. ontories) at the seaward entrance that serve as break- Þá bjó Önundr at Mosfelli. Hallfreðr átti at waters. In this anchorage, ships could safely wait out [gjalda] hálfa mörk silfrs húskarli Önundar winter storms and load cargo and passengers. As the ok svaraði heldr harðliga. Kom húskarlinn saga suggests, Leiruvogur Harbor was a proper place heim ok sagði sín vandrædi. Hrafn kvað slíks to keep a ship anchored for resale. And the Mos- ván, at hann myndi lægra hlut bera í þeira fell chiefs profited from the money and goods that skiptum. Ok um morguninn eptir reið Hrafn changed hands at the port. til skips ok ætlaði at höggva strengina ok How important was the port to the Mosfell stöðva brottferð þeira Hallfreðar. Síðan áttu chiefs? From the archaeological record alone, this menn hlut í at sætta þá, ok var goldit hálfu would have been difficult to determine. But the sagas meira en húskarl átti, ok skilðu at því (Sveins- record the readiness of the Mosfell chiefs to defend son 1939, 196). their economic interests in the port. The passage below from Hallfred’s Saga speaks of the warrior Having medieval narrative sources, such as those Hrafn, the son of the Mosfell chief Önund, as ready connected with the Mosfell Valley, or any written to fight to uphold the family’s right to collect port sources at all, is exceptional in Viking archaeology. landing fees. Hallfred, a Viking warrior who arrives While extensive Viking Age sites are found through- by ship from Norway, refuses to pay to Önund’s out mainland Scandinavia, the British Isles, and servant the landing dues of half a mark of silver, northern Europe, there is a paucity of written sources effectively challenging the chieftain’s right to collect in these regions, so archaeologists, historians, and payments. Hrafn rides to the port and threatens Hall- anthropologists often know little about the inhabit- fred, who backs down and agrees to pay not only the ants, their personal histories, or their specific socio- usual port toll but a humiliating supplement as well. economic and political relationships. It is hard to imagine now, in light of the rich archaeological finds in the Mosfell Valley, that at

134 | backdirt 2013 research in action

Figure 13. Map depicting the Leiruvogur harbor area. Geophysical and oceanographic surveys were conducted to explore the site. This work is being conducted with a team of geophysicists from the University of Kiel, led by Wolfgang Rabbel, Denis Wilken, and Tina Wunderlich, who have joined our project. the start of our excavations many archaeologists, towns and on farms. Our goal was to find sites that historians, and saga scholars thought it was futile to have not been exposed by erosion or construction consult the family sagas as sources for locating sites.7 projects. To this end, we searched Iceland’s medieval We take a contrasting view that, because archaeology texts, including the sagas, for references that could depends on site discovery, MAP employs—with cau- lead to specific sites. The penultimate chapter in tion—every tool and clue that could help us with this Egil’s Saga is a case in point. It names Hrísbrú as the discovery. Most sites in Iceland have been found acci- site where a conversion-period church was built in dentally, when they were exposed by wind or water the Mosfell Valley. The saga also supplies informa- erosion. A large number of sites have been uncov- tion about when, why, and by whom the church was ered during road work and construction, both in built. The following passage from Egil’s Saga led us to the site.

7 Skepticism about the sagas is in part a political legacy. The mid- twentieth-century reinterpretation of the family sagas as thirteenth-century fictional creations was proposed by a group of Icelandic literary scholars known as the Icelandic School (íslenski skólinn). This group emerged at the climax of Iceland’s struggle for independence from Denmark, which Ice- land declared unilaterally 1944, and their theory became institutionalized in the Icelandic educational system. It was and still is the accepted theoreti- cal position among many researchers, particularly archaeologists.

backdirt 2013 | 135 Grím of Mosfell [the chieftain at Mosfell and law (Tulinius 2004, Introduction). While this may husband of Thordís, Egil’s stepdaughter] was be true, Christian law and practice are not always the baptized when Christianity was adopted by same, and discrepancies were particularly common law in Iceland; he had a church built there. in the decades immediately following Iceland’s con- People say that Thórdís had Egil’s bones version to Christianity. The Icelandic narratives of moved to the church, and this is the evidence. the conversion period detail the tolerance of contin- Later when a church was built at Mosfell and ued pagan practices on private farms. It is difficult to that church which Grímr had built at Hrísbrú believe that the first generation of converts to Chris- was taken down, then the graveyard there tianity in Iceland would have adhered closely to the was dug. And under the place of the altar, complicated details of Christian law even if they had human bones were found; they were much known what they were. In any case, Egil was entitled bigger than the bones of other men. People to burial in hallowed ground: during his service as a knew because of the accounts of old men that mercenary for the English king Athelstan (895–933), these were Egil’s bones. At the time, Skapti we are told, he had received prímsigning. Prímsign- the Priest Thórarinsson, a wise man, was ing is a Norse term meaning “provisional baptism,” there. adopted from the Latin primum signum or prima signatio (Byock 1993, 30; see also Molland 1968). (Byock trans.) Prima signatio consisted of making the sign of the cross over non-Christians in order to cleanse them Grímr at Mosfelli var skírðr, þá er kristni var of the evil spirit. After being “prime signed,” pagans í lög leidd á Íslandi; hann lét þar kirkju gera. could attend mass and enter into full relationships En þat er sögn manna, at Þórdís hafi látit with Christians.8 flytja Egil til kirkju, ok erþ að til jarðtegna, The saga also tells us what happened to the early at síðan er kirkja var gör at Mosfelli, en ofan church that Grím built at Hrísbrú: it was “taken tekin at Hrísbrú sú kirkja, er Grímr hafði down” (ofan tekin) or dismantled and moved. We gera látít, þá var þar graf- inn kirkjugarðr. found it 500 meters further eastward up the valley En undir altarisstaðnum, þá fundusk man- toward New Mosfell.9 Furthermore the passage tells nabein; þau váru miklu meiri en annarra us when this occurred: the church was “taken down” manna bein. Þykkjask menn þat vita af sögn while the priest Skapti Thórarinsson was present. gamalla manna, at mundi verit hafa bein From other sources, Prestatal10 and The Saga of Egils. Þar var þá Skapti prestr Þórarinsson, Thorgils and Hafliði,11 we know that Skapti was active vitr maðr (Nordal 1933, ch. 68). in the years 1150–60. If we count the years between This passage gives considerable information and approximately 1155, when the church was taken answers many questions of interest to archaeolo- down, and approximately 1220, when most saga gists. It tells us that a Viking Age church was to be scholars agree that Egil’s Saga was written, this time found on Grím’s farmstead at Hrísbrú in the Mosfell Valley; that it was built when Christianity was accepted into law (ca. C.E. 1000) by Grím Svertings- 8 Jón Steffensen considered the issue of the transference of Egil’s son, the chieftain at Mosfell, because he converted remains to sanctified ground in light of later twelfth- or thirteenth-century to Christianity; that the church included a burial regulations as preserved in Grágás. He argues that, in the early years after ground containing the remains of the warrior poet the conversion, Egil was eligible for reburial in Grím’s new church; see Steffensen 1975, 153. See also Byock 2001, chs. 15 and 18. Egil Skalla-Grímsson; that his remains were moved 9 In 1995, we excavated a corner of what appeared to be the twelfth- there by Thórdís. The sources for this information century church at Mosfell. See Earle et al. 1995. are also given: “people say” (sögn manna) and “the 10 “Nafnaskrá íslenzkra presta,” in Diplomatarium Islandicum: Íslenzkt fornbréfasafn, vol. 1, edited by Jón Sigur sson (Copenhagen: S. L. Möller and accounts of old men” (sögn gamalla manna)—that is, ð Hið íslenzka bókmentafélag, 1857), 186. oral memory. 11 Skapti the Priest participated in feuds. In The Saga of Thorgils and Was there a hindrance to pagan Egil being Hafliði, he is credited with the famous statement: “Costly would be all of reburied in a Christian context? It has been argued Hafliði, if this should be the price of each limb” (D´yrr myndi Hafliði allr, ef svá skyldi hverr limr). The statement refers to the large sum demanded by that a reburial such as this went against Christian Hafliði for the loss of a finger Þ( orgils saga ok Hafliða, ch. 31, in Jóhannes- son et al. 1946).

136 | backdirt 2013 research in action

span was within the memory of one long lifetime. If we count years from the building of the church around 1000 to the dismantling of the church around 1155, this period would be within the memory of two or three long lives. Given the stability of the settlement pattern, the visibility of the church site in the Mosfell Valley both before and after the abandonment, the memory of the site retained in the local place names, and the importance of Egil as Iceland’s great warrior-poet, it is hard to imagine that the story of the church would have been forgot- ten. It is worth noting, too, that Egil’s burial was not a folkloric event claimed by numerous places. There are no other traditions of Egil being buried elsewhere, not even in his ancestral seat at Borg in Borgarfjörður. The saga thus led us to the site, but the identi- fication of the thousand-year-old graveyard was a separate archaeological question. We first turned to geophysical testing of Kirkjuhóll and the homefield Figure 14. The reburied skeleton in Burial Feature 4 was found directly to the north, but the resulting magnetometer lying up against the southern foundation wall of the church. This and resistivity maps yielded negative results and did individual, who may have suffered from tuberculosis and a fatal not suggest the presence of subterranean archi- brain infection, was moved from a pagan grave into the Christian tectural features. Nevertheless because of its place graveyard after the conversion of Iceland to Christianity. name, we decided Kirkjuhóll (Church Knoll) was worth testing with excavations. Once the excavations violent life (Walker et al. 2012). Several individuals began, we soon found domestic refuse from a Viking in the Hrísbrú cemetery show evidence of strenu- Age farm. Then we found a concentration of graves, ous physical activity involving the hands and arms, all with the east–west orientation of Christian burial, and osteoarthritis was prevalent (Eng in press). The indicating the presence of a churchyard. Next, the skeletons also show signs of infectious diseases. One excavation revealed the foundations of a small build- young man has lesions on the pleural surfaces of his ing amid the graves (see Fig. 6; see Byock 2009 for ribs and another young male’s skull shows evidence issues surrounding discovery of the Hrísbrú church of lesions associated with chronic ear infection that and the first phases of the excavation). The skeletal resulted in a brain abscess (Fig. 14). The lesions in remains excavated at Hrísbrú offer considerable evi- both cases suggest that tuberculosis was present in dence about the health status and living conditions the Hrísbrú population (Holck in press). Our data of the tenth- and eleventh-century inhabitants of the show that stressful living conditions and heavy labor Hrísbrú farmstead (Walker et al. 2004). were common among early Icelanders even at a At this point, it was clear to us that archaeology high-status site such as Hrísbrú. Together the dif- and sagas complemented each other. The texts and ferent sources are giving us a broad-based picture of archaeology support each other in illuminating the life on the Hrísbrú farm. economic life of these Viking Age people centered Traumatic injuries have also been found in the on a settled pastoral life of livestock-raising, coastal Hrísbrú cemetery. One person is a homicide victim fishing, and the gathering of wild foods in a chal- with two massive head injuries caused by axe or lenging marginal environment (Byock 2001, 43–62, sword (Fig. 15; Walker et al. 2012). Such evidence of Zori et al. 2013). The texts helped us to discover the graveyard at Hrísbrú that proved rich in bioarchaeological information concerning the Viking Age Icelandic life. From skeletal analysis, we have been able to document a rough and sometimes

backdirt 2013 | 137 Figure 15. The axed-man of Mosfell found in the Hrísbrú graveyard. This man, in his mid-forties, was found close to the eastern wall of the church chancel (Feature 2 in Fig. 7). Reflecting the violence of feuding described in the sagas, he died of head wounds from an axe or sword. Radiocarbon dating places this man’s death in the latter half of the tenth century or the early part of the eleventh.

lethal violence at Hrísbrú is consistent with the gen- archaeology? While we do not by any means believe eral picture of Viking Age Iceland’s feuding society everything found in the written materials, the Icelan- sketched in the sagas (Byock 1982; 2001). The kill- dic sources concerning Mosfell are often informa- ing at Hrísbrú has a parallel in an account from The tive, detailed, and worthy of consideration. They Saga of Gunnlaug Serpent-Tongue. That saga describes offer core information about settlers, chieftains, the violent culmination of the Mosfell chief Önund’s warriors, lawgivers, slaves, and travelers in the feud with the chief Illugi the Black from Gilsbakki Mosfell Valley and the port at Leiruvogur, shedding and the attack on Önund’s farmstead in the Mosfell light on the material culture, social conditions, and Valley. According to the saga: site location. They provide details about things such as the interiors of habitation sites, kinship relations, It is said, that in the autumn Illugi rode mortuary customs, and economic arrangements. from his home at Gilsbakki with thirty men Iceland’s medieval writings comprise northern and arrived at Mosfell early in the morning. Europe’s most comprehensive portrayal of a func- Önund and his sons escaped into the church, tioning medieval society, and MAP is using them but Illugi caught two of Önund’s kinsmen, to develop a novel methodology for archaeological, one named Björn and the other Thórgrím. anthropological, and historical research (Byock 2001, Illugi had Björn killed and Thórgrím’s foot 21–24 and 149–51; Byock 1994b). Taken together, chopped off. Then he rode home and after MAP’s findings confirm the value of this multidisci- this Önund sought no reprisal.12 plinary approach to the sources for the study of early What conclusions can we draw at this stage Iceland and the Viking Age. X of the Mosfell excavations concerning sagas and

12 Gunnlaugs saga ormstungu, ch. 13, in Borgfirðinga sögur: Íslenzk fornrit 3, edited by S. Nordal and G. Jónsson (Reykjavík: Hið Íslenzka Fornritafélag, 1938).

138 | backdirt 2013 research in action

Figure 16. Excavating the Viking Age longhouse at Hrísbrú. The stones being exposed here made up the inside of the longhouse walls. In this photograph the collapsed turf roof and side walls still cover the inside features and floors of the house.

References

Bathurst, Rhonda, Davide Zori, and Jesse Byock. 2010. Diatoms ———. 2001. Viking Age Iceland. London: Penguin Books. as Bioindicators of Site Use: Locating Turf Structures ———. 2009. Findings from the Mosfell Archaeological Project’s from the Viking Age. Journal of Archaeological Science 37: Seminal 2002 Excavations. In Heimtur: Ritgerðir til heiðurs 2920–28. Gunnari Karlssyni sjötugum, edited by Helgi Skúli Kjartans- Byock, Jesse. 1982. Feud in the Icelandic Saga. Berkeley: University son Guðmundur Jónson, Vésteinn Ólason. Reykjavík: Mál of California Press. og menning.

———. 1993. The Skull and Bones in Egil’s Saga: A Viking, A ———. In press. The Mosfell Archaeological Project: Archae- Grave, and Paget’s Disease. Viator: A Journal of Medieval ology, Sagas and History. In Viking Age Archaeology in and Renaissance Studies 24: 23–50. Iceland: the Mosfell Archaeological Project, edited by Davide Zori and Jesse Byock. Turnhout: Brepols Publishers. ———. 1994a. Hauskúpan og beinin í Egils sögu. Skírnir: Tímarit hins íslenska bókmenntafélags: 73–109. ———, Phillip Walker, Jon Erlandson, Per Holck, Davide Zori, Magnús Guðmundsson, and Mark Tveskov. 2005. A ———. 1994b. Social Memory and the Sagas: The Case of Egils Viking-Age Valley in Iceland: The Mosfell Archaeologi- saga. Scandinavian Studies 76: 299–316. cal Project. Medieval Archaeology, 49: 195–218. [Icelandic: ———. 1995. Egil’s Bones: A Viking Warrior and Paget’s Disease. idem. 2007. Valdimiðstöð í Mosfellsdal: Rannsóknir á Scientific American 272: 82–87. [German: Die Egil-saga fornleifum frá tímum Víkinga að Hrísbrú og Mosfelli. und das Paget-Syndrom. Spectrum der Wissenschaft (March Ólafia: Rit fornleifafræðingafélags, 2: 84–106.] 1995); French: Les os d’Egil, héros viking. Pour la Science ———, and Davide Zori. In press. Introduction: Viking Archaeol- 209 (March 1995): 52–58; Italian: Le ossa di Egill. Le Sci- ogy and the Mosfell Archaeological Project. In Viking Age enze 319 (March 1995): 74–79; Polish: Ko´sci Egila. Swiat´ Archaeology in Iceland: the Mosfell Archaeological Project, Nauki (March 1995): 72–77.] Also available in Archaeology: edited by Davide Zori and Jesse Byock. Turnhout: Brepols Annual Edition 1996/97 (Guilford: Brown and Benchmark Publishers. Publishers): 80–85.

backdirt 2013 | 139 Figure 17. Part of the MAP team at the end of the 2007 season. Andrés Ólafsson, the farmer at Hrísbrú, is on the far left.

Connors, Colin G. In press. Viking Age Routes, Landscape, and Grimes, Vaughan, Benjamin T. Fuller, and Eric J. Guiry. In press. Power in the Mosfell Region. In Viking Archaeology in Reconstructing Diets and Origins of Vikings at Hrís- Iceland: The Mosfell Archaeological Project, edited by Davide brú, Mosfell Valley, Iceland: the Carbon, Nitrogen and Zori and Jesse Byock. Turnhout: Brepols Publishers. Strontium Isotope Evidence. In Viking Age Archaeology in Iceland: the Mosfell Archaeological Project, edited by Davide Earle, Timothy, Jesse Byock, Phillip Walker, and Sigurður Berg- Zori and Jesse Byock. Brepols Publishers, Turnhout. steinsson. 1995. Mosfell Archaeological Project 1995 Field Season. Reykjavík: Unpublished report submitted to Forn- Hansen, Sigrid Cecilie Juel, Davide Zori, and Jesse Byock. In leifavernd Ríkisins. press. Artefacts from the Viking Age Longhouse at Hrísbrú. In Viking Age Archaeology in Iceland: the Mosfell Eng, Jacqueline T. In press. Bioarchaeological Evidence of Stress Archaeological Project, edited by Davide Zori and Jesse and Activity in the Medieval Hrísbrú Cemetery in the Mos- Byock. Turnhout: Brepols Publishers. fell Valley, Iceland. In Viking Age Archaeology in Iceland: the Mosfell Archaeological Project, edited by Davide Zori and Hilberg, Volker, and Sven Kalmring. In press. Viking Age Hedeby Jesse Byock. Turnhout: Brepols Publishers. and Its Relations with Iceland and the North Atlantic: Communication, Long-distance Trade and Production. In Erlandson, Jon M., Jesse Byock, and Davide Zori. In press. Egill’s Viking Archaeology in Iceland: The Mosfell Archaeological Grave? Archaeology and Egils Saga at Kirkjuhóll, Hrísbrú. Project, edited by Davide Zori and Jesse Byock. Turnhout: In Viking Archaeology in Iceland: The Mosfell Archaeological Brepols Publishers. Project, edited by Davide Zori and Jesse Byock. Turnhout: Brepols Publishers. Holck, Per. 2005. Egill Skallagrimssons gård og kirke på Island - fra utgravningene 2001–2005. Michael 2 (2005), 340–48. Fríðriksson, Adolf. 2009. Social and Symbolic Landscapes in Late Iron Age Iceland. Archaeologia Islandica 7: 9–21. ———. In press. Infectious Disease in Viking Age Iceland. In Viking Archaeology in Iceland: The Mosfell Archaeological Project, edited by Davide Zori and Jesse Byock. Turnhout: Brepols Publishers.

140 | backdirt 2013 research in action

Jóhannesson, Jón, Magnús Finnbogason, and Kristján Eldjárn, ———. In press. Interdisciplinary Modelling of Viking Age eds. 1946. Sturlunga Saga 1. Reykjavík: H. F. Leiftur. and Medieval Settlement in the Mosfell Valley. In Viking Archaeology in Iceland: The Mosfell Archaeological Project, Milek, Karen, Davide Zori, Colin Connors, Waltraud Baier, Kate edited by Davide Zori and Jesse Byock. Turnhout: Brepols Baker, and Jesse Byock. In press. The Organization and Publishers. Use of Space in the House at Hrísbrú. In Viking Age Archaeology in Iceland: the Mosfell Archaeological Project, ———, and Jesse Byock, eds. in press 2014. Viking Archaeology edited by Davide Zori and Jesse Byock. Turnhout: Brepols in Iceland: The Mosfell Archaeological Project. Turnhout: Publishers. Brepols Publishers.

Molland, Einar. 1968. Prímsigning. Kulturhistorisk leksikon for ———, Thomas Wake, Jon M. Erlandson, and Rúnar Leifsson. nordisk middelalder 13. Reykjavík: 439–44. In press. Viking Age Foodways at Hrísbrú. In Viking Archaeology in Iceland: The Mosfell Archaeological Project, Nordal, Sigurður, ed. 1933. Egils saga Skalla-Grímssonar: Íslenzk edited by Davide Zori and Jesse Byock. Turnhout: Brepols fornrit 2. Reykjavík: Hið Íslenzka Fornritafélag. Publishers. Steffensen, Jón, ed. 1975. Ákvæði kristinna laga þáttar um bein- afærslu. In Menning og Meinsemdir: Ritgerðasafn um mótu- narsögu íslenskrar þjóðar og baráttu hennar við hungur og sóttir. Reykjavík.

Sveinsson, Einar Ól., ed. 1939. Vatnsdæla saga, Hallfreðar saga, Kormáks saga: Íslenzk fornrit VIII. Reykjavík: Hið Íslenzka Fornritafélag.

Thórðarson, Björn Thráinn. In press. Government’s Partnership with the Mosfell Archaeological Project: Past, Present and Future. In Viking Archaeology in Iceland: The Mosfell Archae- ological Project, edited by Davide Zori and Jesse Byock. Turnhout: Brepols Publishers.

Tulinius, Torfi. 2004. Skáldið í Skriftinni: Snorri Sturluson og Egils saga. Reykjavík: Hið Íslenska Bókmenntafélag. Walker, Phillip, Jesse Byock, Jacqueline Eng, Jon Erlandson, Per Holck, Kaethin Prizer, and Mark Tveskov. 2004. Bioar- chaeological Evidence for the Health Status of an Early Icelandic Population. Paper presented at the 73rd meeting of the American Association of Physical Anthropologists, Tampa, FL.

Walker, Phillip, Jesse Byock, Jon Erlandson, Per Holck, Jacqueline Eng, Henry Schwarz, Davide Zori. 2012. The Axed Man of Mosfell: Skeletal Evidence of a Viking Age Homicide and the Icelandic Sagas. In The Bioarchaeology of Individuals, edited by Ann Stodder and Ann Palkovich. Gainesville: University of Florida Press.

Zori, Davide. 2007. Nails, Rivets, and Clench Bolts: A Case for Typological Clarity. Archaeologia Islandica, 6: 32–47.

———. 2010. From Viking Chiefdoms to Medieval State in Ice- land: The Evolution of Power Structures in the Mosfell Valley. Ph.D. dissertation, UCLA.

———, Jesse Byock, Egill Erlendsson, Steve Martin, Thomas Wake, and Kevin J. Edwards. 2013. Feasting in Viking Age Iceland: Sustaining a Chiefly Political Economy in a Mar- ginal Environment. Antiquity 87 (335): 150–65.

backdirt 2013 | 141